Evaluating Responses to Conflict No.1/September 2009 Policy Brief Evaluating Responses to Conflict Conflict and Development Program 53335 Edition I September 2009 Reintegration and Localized Conflict: Promoting Police-Combatant Communication Worldwide, reintegration programs have IntroductIon gained wide recognition as an important This briefing is based on the following element of post-conflict peace-building. By Despite frequent application internationally, article: providing economic assistance to combat- disarmament, demobilization and reintegra- ants and helping them to gain social accep- tion (DDR) programs are new to Indonesia. Dave McRae (2009). "DDR and tance, programs hope to reduce the chance Formal DDR programs were first attempted Localized Violent Conflict: Evaluating of ex-combatants acting as spoilers. Its in Indonesia in 2005 in Aceh and then for a Combatant Reintegration Programs second time in 2007-2008 in Poso district, widespread application notwithstanding, in Poso, Indonesia". Indonesian Social Central Sulawesi. Unlike the Aceh programs, the empirical basis to judge the effective- Development Paper No. 14. Jakarta: ness of reintegration approaches remains DDR in Poso involved only reintegration ­ World Bank. limited, as does our understanding of the there was no disarmament or demobiliza- mechanisms by which reintegration pro- tion component. grams produce security impacts. In its integrated DDR standards, the United This note presents the findings of a Conflict Nations defines reintegration as "the pro- and Development (C&D) program study cess by which ex-combatants acquire civil- Two series of Policy Briefs are that evaluates two combatant reintegration ian status and gain sustainable employment published by the Conflict and programs undertaken in Poso district, Cen- and income" (UN 2006). Beyond this basic Development program within the tral Sulawesi, Indonesia from 2007-2008. definition, reintegration programs differ World Bank Indonesia country The Poso programs are interesting because in approach and scope. Programs may se- they involved the application of conven- lect combatants individually as recipients team. The first, "Understanding tional reintegration approaches in a novel (individually-targeted) or may ask commu- Conflict Dynamics and Impacts," nities to determine allotment of resources context of localized, sporadic violence. summarizes the results of research (community-driven). Programs may focus on conflict in Indonesia. These briefs on short-term security goals (minimalist) or The study finds that the primary value of re- draw out lessons for understanding integration in Poso was its role in assisting be explicitly oriented towards broader post- and responding to conflict in other authorities to establish relationships with conflict recovery (maximalist) (Kingma and middle-income countries. The combatants which in turn helped them bet- Muggah 2009). second, "Evaluating Responses to ter manage security. Police increased their Conflict," summarizes the results levels of contact with combatants through The Poso programs were an example of of evaluations and assessments reintegration and other informal economic individually-targeted minimalist reintegra- of conflict programs, and the inducements, and were able to leverage tion. There were two programs in Poso: a this contact to gather information in the af- police-designed vocational training program methodological implications of how termath of security incidents and to detect and a cash payment program run by the lo- we can best measure the impacts of cal government. Introduced shortly after potential security disturbances. peace-building programs. All Policy police raids halted violence in Poso in 2007, Briefs are available at each program aimed to prevent further at- There are two implications for reintegra- www.conflictanddevelopment.org tion theory and practice. If reintegration tacks by addressing perceived economic dif- is understood as a means to provide in- ficulties experienced by youths whose main ducements to cooperate with police, then skill was perpetrating violence (Karnavian individual targeting is crucial. In light of et. al. 2008, 377). If combatants were occu- this, planners need to limit the resources pied with "productive concerns", planners allocated for reintegration to a level that hoped, they would be less likely to engage achieves security goals but minimizes the in renewed violence that could endanger longer-term negative impacts associated their businesses. Police also hoped that the with prioritizing ex-combatants. programs would provide a forum for them Policy Brief | Conflict and Development Program - World Bank Office Indonesia 1 Evaluating Responses to Conflict No.1/September 2009 to establish relationships with com- largely confined to just one of Indo- view. Groups of respondents with par- batants. In addition to the formal pro- nesia's 400-odd districts. This local ticular characteristics could be identi- grams, the authorities also provided scale meant that the programs faced fied from the quantitative data, after combatants with other preferential a caseload of hundreds rather than which the relevant qualitative tran- access to resources, such as construc- tens of thousands of combatants, scripts were used for process-tracing. tion projects allocated without tender making trade-offs between coverage and free administrative documents. and quality of assistance less neces- Key informant interviews were con- sary. Additionally, social acceptance of ducted with program planners and Several atypical features of the Poso combatants was not a major problem implementers, other police and gov- conflict may have been expected to in Poso, as combatants typically lived ernment officials, and with combat- aid the effectiveness of the reintegra- within their own communities while ants who played a role in selecting tion programs. As serious as violence perpetrating violence. program participants. Focus group dis- in Poso has been, the conflict was cussions and key informant interviews But the localized pattern of violence in were also held to investigate commu- The Poso conflict Poso also meant that having a job was nity attitudes to the programs. not fundamentally incompatible with The Poso conflict was the most taking part in violence. Indeed, the Two aspects of the security impacts of persistent of several communal majority of program participants ap- each program were assessed: direct conflicts that marked Indonesia's pear to have been engaged in at least effects, namely influences on the mo- post-authoritarian transition. In the some form of economic activity prior tivations and capacity of combatants later years of the conflict, the district to receiving reintegration assistance. to perpetrate further violence; and also became an important arena of indirect effects, namely influences on operations for Indonesia's jihadist The methodology and main findings police capacity to manage security. networks. An estimated 600 to 1,000 of the C&D study are summarized be- people were killed in fighting between low. This note addresses the following Although field research covered all Christians and Muslims in Poso (the key questions: groups of participants, analysis in this state was not a direct party to the violence), more than half of whom note focuses on Muslim male par- met their deaths in 2000 and 2001. · Is individually-targeted reintegra- ticipants. Muslim males were by far Following the peak years of fighting, tion an appropriate strategy to the largest group of program partici- the conflict entered a protracted phase contribute to improved security in pants, and it was with them in mind of sporadic violence, during which situations of localized conflict such that planners formulated the strategic violence became increasingly one-sided, as Poso? aims of reintegration in Poso. as a core group of Muslim combatants · Through what mechanisms can re- targeted non-Muslims. This sporadic integration assistance be effective FIndIngs violence was finally halted in early 2007 after localized conflict? · What costs are inherent in attempt- Contrary to the assumption of pro- by two major police raids. gram planners of widespread ex- ing individually targeted reintegra- tion under these circumstances? combatant unemployment, most Combatant entities in Poso were combatants interviewed reported clandestine and were never rigidly Methodology that they were engaged in economic structured, meaning it is often difficult to identify who was or was not a activity immediately prior to taking combatant. Indeed, "combatant" To investigate program impacts, part in the programs. Only around is used here as a shorthand for around one-sixth of program partici- a quarter of program participants perpetrators or active supporters of pants were selected as respondents interviewed said that they were un- violence - the term is not widely used via stratified random sampling proce- employed or underemployed at the in Poso. dures. Each respondent completed a beginning of 2007, shortly before the quantitative questionnaire followed programs began. This finding calls by a semi-structured qualitative inter- into question the programs' rationale: Table 1: Reintegration programs in Poso if most combatants were employed even prior to the programs, then ex- Non- Program Form of Assistance Budget Program Combatant combatant combatant idleness is less likely to duration recipients recipients have been as significant a security Yes threat as program planners assumed. Individual-targeted Rp. 1.7 billion July 2007; 227 Muslim (female Police (training and (US$ Nov 2007 ­ (two phases), training and There are two necessary caveats. Ex- -designed in-kind enterprise 170,000, May 2008 20 Christian capital) approximate) (two phases) (approx.) micro-credit combatants may have incorrectly re- groups) called their employment status. Their Individual-targeted willingness to engage in vocational (cash payment to Rp. 2 billion Yes training also suggests that whatever Poso each individual, 172 Muslim, (wives, family their prior employment status was, (US$ District Rp. 10 million to Aug 2007 87 Christian of men Gov. Muslims, 200,000, (approx.) arrested or they were keen to increase their level approximate) Rp 2.5 million to killed) of economic activity. Christians) 2 Policy Brief | Conflict and Development Program - World Bank Office Indonesia Evaluating Responses to Conflict No.1/September 2009 The programs were unlikely to have Map 1: Poso district including invitations to been the main factor exerting a direct combatants to participate influence on security improvements in religious study sessions in Poso. The situation in Poso had im- with the Poso police chief, proved markedly even before the pro- himself a devout Muslim. grams began as a result of the Janu- ary 2007 police raids. Indeed, there The granting of preferen- were no security incidents during the tial access to economic two months immediately prior to the resources also created a commencement of the programs. Be- mutual interest in contact cause security had improved before for police and ex-combat- the programs began, the broader ants. Some ex-combatants question for reintegration practice of reported that they contact whether occupying combatants can the police to access free effectively contribute to security in drivers' licenses or vehicle localized conflict settings remains un- registration. One man had answered. Regardless, the programs even positioned himself produced positive longer-term eco- confrontations from escalating. Po- as an agent for local residents who nomic outcomes for only a fraction of lice also use their contacts to convene needed to arrange these documents. participants. If occupying combatants meetings of ex-combatants ahead of Other said they would make contact was a viable mechanism to contribute periods when security risks are per- with either the police or the govern- to security in Poso, then alterations ceived to be elevated. ment when they sought access to con- to design and implementation of the struction projects. Indeed, almost a programs could have been made to The forms of contact and motiva- quarter of Muslim ex-combatants in- increase the period during which they tions for respondents to commu- terviewed mentioned the importance occupied combatants. nicate with police vary widely. Not of construction contracts to their live- all contact between police and ex- lihoods or to the security situation. The clearest program effect on secu- combatants takes place on voluntary rity was achieved through increased terms - police forcing the issue is part Granting targeted reintegration as- levels of contact between police and of the picture. Indeed, some ex-com- sistance and preferential access to Muslim ex-combatants. Establishing batants expressed resentment over resources to ex-combatants in Poso relations with ex-combatants in Poso contact with police that more closely has generated negative side-effects, constituted a formidable challenge resembles surveillance than recipro- but the extent to which these costs for police due to widespread hostility cal communication. Such resentment have manifested has been limited by generated by past failures in law en- does not mean police are wrong to the relatively modest commitment forcement and jihadist doctrine that engage in surveillance or to take the of resources thus far. Between them police were thoghut (anti-Islamic) initiative in contacting ex-combatants the two reintegration programs cost (ICG 2005, 2007). Furthermore, po- to gather information, but it does re- around Rp. 4 billion (US$400,000), lice had killed sixteen Muslim youths mind us that building good relations a fraction of the post-conflict aid al- during police raids in 2006 and 2007, is not the only tool that police use to located to Poso. Most construction creating a fresh source of anger. Nev- manage security. contracts allocated to ex-combatants ertheless, almost half of Muslim re- also appear to have been of relatively spondents reported being in contact Other contact appears to derive from modest value. This level of assistance with police, with some respondents the programs and a broader police does not appear to have entrenched saying they would not have been will- strategy of provid- ing to communicate with police prior ing preferential ac- Figure 1: Reported contact with police by arrested status, to taking part in the programs. cess to resources Muslim respondents to combatants, Police are able to leverage contact called the "persua- with ex-combatants to gather infor- sive" approach. mation in the aftermath of security The programs pro- incidents and to detect potential se- vided police with a curity disturbances as they are de- reason to contact veloping. Their relationships with ex-combatants, ex-combatants also serve as a point ostensibly to ask of contact for police to manage secu- how their business rity when security incidents do occur. enterprises were When two brawls took place in mid- going. Police have 2007 after ex-combatants objected also made a priority to other youths consuming alcohol, of pursuing contact police were able to contact influential with ex-combatants ex-combatants to try to prevent the through other fora, Policy Brief | Conflict and Development Program - World Bank Office Indonesia 3 Evaluating Responses to Conflict No.1/September 2009 Figure 2: Change in occupation 2007 - 2008/9, Muslim respondents nor costs can be known in advance. But planners should explicitly consid- er the possible negative side-effects of individually-targeted reintegration during the design phase. recommendatIons · Consider individually-targeted rein- tegration immediately after localized conflicts to address hostility and lack of contact between authorities and combatants. · If reintegration is conducted to in- crease police/combatant interac- tion, limit resource allocation to a level that achieves security goals while minimizing negative side- ex-combatants ­ or at least those ex- conclusIons effects. combatants who took part in the pro- · Include investigation of likely lon- grams ­ as dominant players in the The available data suggests that most ger-term costs when evaluating district economy. But the longer ex- ex-combatants in Poso did not face reintegration programs. combatants are provided with prefer- particular obstacles in accessing work ential access to resources, the more or economic resources when the con- PaPers cIted they are likely to consolidate their flict came to an end, a pattern also Baare, Anton (2006). "An Analysis of Transitional economic position and to develop a sense of observed in the aftermath of the Aceh Reintegration", Background Paper for the Swedish Initia- entitlement. The sooner preferential conflict (Barron 2009). Indonesia may tive for Disarmament, Demobilization and Reintegration access to resources is ended or scaled thus present a set of reintegration (SIDDR). down in Poso, the more manageable cases where one of the main conven- Barron, Patrick (2009). "The Limits of DDR: Reintegration possible associated difficulties such as tional rationales for targeted reinte- Lessons from Aceh" in Small Arms Survey Yearbook 2009. low level attacks and intimidation are gration ­ assisting ex-combatants to Cambridge: Cambridge University Press. likely to be. overcome acute reintegration chal- International Crisis Group (2007). "Deradicalisation" and lenges (Tajima 2009) ­ does not apply. Indonesian Prisons, Asia Report no. 142. Jakarta/Brussels: If reintegration is understood as an Under such circumstances, the main International Crisis Group. intervention to increase contact be- value in targeted reintegration may lie ________ (2005). Weakening Indonesia's Mujahidin Net- tween authorities and ex-combat- in helping authorities to establish re- works. Asia Report no. 103. Jakarta/Brussels: International ants, then an individually-targeted lationships with ex-combatants to en- Crisis Group. approach is likely to be most suit- able them to better manage security. Jennings, Kathleen (2008). "Unclear Ends, Unclear Means: able. Community-driven programs Reintegration in Postwar Societies ­ The Case of Liberia". typically focus on public goods mak- In highlighting the potential security Global Governance 14: 327-345. ing them less suited to provide indi- impacts of reintegration measures, vidual inducements to cooperate with we must also consider the ethical and Karnavian, M. Tito et al. (2008). Indonesian Top Secret. Ja- karta: Gramedia Pustaka Utama. police. Implementers of individually- developmental costs. When entire target reintegration may also be able communities have been affected by Kingma, Kees and Robert Muggah (2009). "Critical Issues to narrowly target recipients, thereby conflict, the prioritization of combat- in DDR: Context, indicators, targeting, and challenges", Background paper for the International Congress on Dis- limiting the level of funding required ants over others can be seen as an armament, Demobilization and Reintegration (CIDDR), to run programs. Individual targeting ethical trade-off between security Colombia. may thus assist program planners to benefits and the fair allocation of re- Tajima, Yuhki (2009). "Background Paper on Economic Re- determine a level of resource alloca- sources (Baare 2005, Jennings 2008). integration", Background paper for CIDDR, Colombia. tion that achieves security goals but Offering significant incentives to com- minimizes the longer-term negative batants may also risk undermining Torjesen, Stina (2006). "The political economy of Disar- impacts associated with prioritiz- longer-term development, if doing so mament, Demobilisation and Reintegration (DDR)", Norsk Utenrikspolitisk Institutt Paper no. 709. Oslo: NUPI. ing ex-combatants for assistance in a entrenches combatants' "economic post-conflict environment. and political standing" (Torjesen United Nations (2006). Integrated Disarmament, Demobi- 2006,7). Of course neither benefits lization and Reintegration Standards (IDDRS). For further information, please contact: Conflict and Development Team Jl. Peurada Utama No. 11A, PSF - The World Bank Satellite Office Gampong Peurada, Banda Aceh info@conflictanddevelopment.org Jl. Diponegoro No. 72, Menteng Phone: +62-651 755 1176 Jakarta 10310 - INDONESIA Fax: +62-651 755 1178 Phone: +62-21-314 8175 Fax: +62-21-3190 3090 Please visit our website: email: info@conflictanddevelopment.org http://www.conflictanddevelopment.org 4 Policy Brief | Conflict and Development Program - World Bank Office Indonesia